lte small cell v.s. wifi user experience · smoother online video experience. (a) lte small cell...
TRANSCRIPT
LTE Small Cell v.s. WiFi User Experience
1. Introduction ..................................................................................................................2
2. Performance Analysis and Evaluation ............................................................................3
2.1 User Experience Test of Typical Services ........................................................................ 3
2.1.1 Test Environment & Scenarios .............................................................................. 3
2.1.2 QoS Test of Typical Services ................................................................................. 4
1) Ping latency ............................................................................................................. 4
2) FTP downloading time.............................................................................................. 5
3) Online Video Service................................................................................................. 5
2.1.3 Robustness Test of Cell Throughput versus Number of Users ............................... 6
2.2 Link-adaptation Mechanism Comparison ...................................................................... 7
2.3 Summary ...................................................................................................................... 8
3. Standard Evolution and Industrial Environment .............................................................9
3.1 Standard Evolution ....................................................................................................... 9
3.1.1 Standard Evolution Roadmap of LTE Small Cell .................................................... 9
3.1.2 Standard Evolution of WiFi .................................................................................. 9
3.2 Industry Ecosystem Analysis ....................................................................................... 10
3.2.1 Deployment Scenarios of LTE Small Cell ............................................................. 10
3.2.2 Deployment scenarios of WiFi ........................................................................... 10
3.3 Commercial maturity analysis ..................................................................................... 11
4. Conclusion ...................................................................................................................12
5. References ...................................................................................................................13
1
LTE Small Cell v.s. WiFi User Experience
Keywords: LTE, WiFi, user experience, standard evolution
Abstract: This document summarizes a set of tests of user experience in
indoor hotspot scenarios with two small cell solutions, LTE Pico
and WiFi. The analyses and tests results show that LTE small cell
has very good stability in providing the satisfied user quality of
experience, especially in high load scenarios. As observed from
the whole set of tests, LTE small cell is more suitable for the
scenario with high UE density as in operator deployed networks
or enterprise networks.
2
1. Introduction
Industry predictions have shown that the future data services will lead to
an exponential data traffic growth by 1000 times[1] [2]. According to these
statistics, 80% of the traffic will be indoor and at hot spots[3]. To meet the
large capacity demand indoor and at hot spots, two popular solutions are
now being discussed: expending the deployment of WiFi, or deploying small
cells based on enhancements of the cellular technology[4]. Both systems
keep evolving in the corresponding standard organizations, e.g. small cell
enhancements for LTE, which are currently being standardized in 3GPP Rel-12
(to be frozen in 2014Q3), are also known as LTE-Hi; as well as IEEE802.11ac
(to be frozen in 2014Q1) for WiFi enhancement. This document provides an
analysis and test results for the user experience of the commercial-level LTE
small cell, as well as a comparison with the other small cell solution, namely
WiFi.
3
2. Performance Analysis and Evaluation
2.1 User Experience Test of Typical Services
2.1.1 Test Environment & Scenarios
The performance comparison is based on LTE macro eNodeB and WiFi Access
Point (AP), where LTE is based on Rel-8 while WiFi is based 802.11n.
The test environment is single cell scenario without inter-cell interference,
with the detailed configurations listed in Table 1. These set of tests focus on
single cell scenario, since WiFi is mainly for indoor isolated deployment thus
it would be hard to test the inter-cell interference mitigation effect without a
centralized WiFi access controller (AC).
Considering that WiFi is a TDD system, we chose TD-LTE small cell in the
comparison of user experience test.
Table-1: Test environment and scenario configuration of LTE small cell and WiFi
*Note: Category 4 type UE can support up to 150Mbps at 20MHz with 2 antennas.
Test environment
configurationLTE small cell WiFi
Scenario Single Cell Single cell
Carrier frequency 2.6GHz 2.4GHz
Tx Power 23dBm 24dBm
NodeB/AP Type TD-LTE eNodeB, based on LTE Rel-8AP: TP-Link router(TL-WR1041N),
based on 802.11n
UE/STA Category 5 Test-UEs (Category 4)*
5 high category STA in total,
where 3STA is laptop, 2STA is
router configured as STA.
uplink-downlink / special
subframe configurations
(8:1:1) / (10:2:2) for Downlink(DL)
(1:1:3) / (10:2:2) for Uplink(UL)
(3:1:1) / (10:2:2) for DL and UL
N/A
Antenna configurationDL: 2x2
UL: 1x2
DL: 2x2
UL: 2x2
2
4
Figure 1 shows that the ping latency of LTE remains stable around 10ms,
while the ping latency of WiFi increases dramatically, from 7ms to 70ms with
increasing number of concurrent FTP users, due to the contention based
resource competition scheme in WiFi.
2.1.2 QoS Test of Typical Services
The user experience is tested for some typical services, including Ping latency,
FTP downloading time and bitrate, online video playing. The tests results are
summarized in the following section.
1) Ping latency
Test metric: Ping latency in milliseconds with an application of uploading
small data packet to a server. Different load scenarios are simulated with 0~4
concurrent FTP users.
Test method:
Send 32kB packet data to server, record ping latency under heavy load 1.
scenario.
The heavy load scenario is such that concurrent FTP users are increased 2.
one by one and one user sends a ping packet.
Figure-1: Ping latency comparison between LTE small cell and WiFi
Ping
Lat
ency
(ms)
LTE small cell
WiFi
0user0
10
20
30
40
50
60
70
80
1user 2user 3user 4user
5
3) Online Video Service
Online movie/video services are becoming more and more popular and will
be the dominant services in the future mobile broad band (MBB) system[5].
The quality of online video service is tested in a scenario with high traffic load
in the same cell, with three video users and two FTP users accessing the cell
simultaneously, as shown in Figure 3.
It is observed that under heavy load scenario with concurrent FTP and online
video users accessing one cell simultaneously, the online video downloading
bitrate of WiFi is lower than that of LTE small cell, and also varies from time
to time which leads to some interruption and discontinuity during the video
playing period. With the same scenario, LTE small cell users have a much
smoother online video experience.
(a) LTE Small cell (b) WiFi
Figure-3: Online Video download rate
2) FTP downloading time
The FTP downloading time is tested in single user and multiple user scenarios,
as shown in Figure 2. It is observed that the FTP download time with LTE
small cell is almost the same in multi-user and single user conditions. For WiFi,
however, the download time increases obviously when the total traffic load
of the WiFi AP reaches a certain level.
Figure-2: FTP Downloading time comparison between LTE small cell and WiFi
Dow
nloa
d Ti
me(
s)
LTE small cell
WiFi
1user0
50
100
150
200
250
300
350
3user 5user 3video+2FTP user
6
Figure-4: FTP DL & UL throughput comparison between LTE small cell and WiFi
Dat
a ra
te (M
bps)
Uplink
Downlink
0
20
40
60
80
100
120
140
It is observed that in order to support the same type of services, the LTE cell
throughput in both downlink and uplink remains constant, independently
of the number of served users. However, the downlink and uplink data
rates of WiFi decrease as the number of served users increases, leading to
unpredictable downlink and uplink throughput per user.
2.1.3 Robustness Test of Cell Throughput versus Number of Users
The cell throughput of both LTE and WiFi are tested for different user density
scenarios, by recording every user FTP uplink and downlink bit rates for the
single cell scenario with different numbers of active users, as shown in Figure 4.
LTE 1user
WiFi 1user
LTE 3user
WiFi 3user
LTE 5user
WiFi 5user
7
2.2 Link-adaptation Mechanism Comparison
LTE and WiFi have different link adaptation and retransmission mechanisms,
according to the specifications:
LTE Rel-8: HARQ + turbo code, with modulation from QPSK to 64QAM•
WiFi IEEE802.11n: ARQ + convolution code, with modulation from QPSK • to 64QAM[6]
Figure-5: Link modulation and coding efficiency comparison between LTE and WiFi, with
corresponding link adaptation mechanisms
HARQ + TC (LTE)
ARQ + CC
Figure 5 shows the relationship between the modulation and coding efficiency
and SINR, for both LTE and WiFi. It is observed that the modulation and
coding efficiencies of LTE is better than WiFi. This is due to the fact that WiFi
does not have dynamic link adaptation, thus the scheduling efficiency varies a
lot for different configurations. Figure 5 shows the potential modulation and
coding efficiency with well-planned initial modulation and coding scheme
selection with ideal channel quality estimation or with 1dB estimation error.
Commercial WiFi routers, however, have quite diverse configurations, thus
some may have even worse efficiency than that in Figure 5.
It is worth noting that both IEEE 802.11ac and LTE Rel.12 introduce higher-
order-modulator up to 256QAM in downlink, which improve the peak
modulation and coding efficiencies to 6.67bits/symbol (WiFi) and 7.43bits/
symbol (LTE), respectively.
8
2.3 Summary
According to the above test and analyses, LTE Small Cell provides good
performance with more efficient scheduling mechanism and better
guaranteed QoS of typical services compared to WiFi:
Ping latency of LTE Small Cell in multi-user scenario is almost the same • as that in single-user scenario, while ping latency of WiFi increases
dramatically with increasing number of served users.
Under heavy load scenario, the FTP downloading time in WiFi is quite • high, and the online video playing is not very smooth with some
interruptions. With the same loaded scenario, LTE Small Cell users
can have a stable FTP downloading time and smoother online video
experience.
The link adaptation mechanism of LTE small cell provides better link • performance than WiFi.
9
3. Standard Evolution and Industrial Environment
3.1 Standard Evolution
3.1.1 Standard Evolution Roadmap of LTE Small Cell
LTE Small Cell keeps evolving in 3GPP, as shown in the LTE standard roadmap
in Figure 6. The latest LTE Small Cell Enhancement (SCE) work items were
setup in Rel-12, and are expected to be completed in June 2014[7][8].
3.1.2 Standard Evolution Roadmap of WiFi
The IEEE802.11 series standard evolution is shown in Figure 7. The core
WiFi functionality standard evolved to 802.11n and 802.11ac based on the
requirement of wider bandwidth and higher data rate for local fixed access.
There is an ongoing study on HEW (high-efficient WLAN) for the potential
further enhancement. Considering the QoS enhancement and new service
support (e.g. video transport stream), other 802.11 standard evolution
branches are introduced, e.g. 802.11ae and 802.11aa.
Figure-6: LTE small cell standard evolution roadmap
Figure-7: IEEE802.11 series standard evolution roadmap
10
3.2 Industry Ecosystem Analysis
3.2.1 Deployment Scenarios of LTE Small Cell
For LTE small cell deployment scenario, the main deployment scenario is in the
cellular network deployed by a mobile network operator, where LTE small cell
can provide the capacity enhancement and coverage extension for macro cell
with the compatible QoS and the efficient coordinated resource management
by taking the advantage of the close coupling between macro cell and small
cell.
There are also scenarios with some hotspot and enterprise small cell coverage
deployed by third party, where LTE small cell can also provide the high
capacity for the coverage of the third party area with the limited access and
authentication, guaranteed security and QoS. The small cell network can
work in a standalone way without interworking with operators’ macro cell.
Recently, the potential LTE small cell operation aggregating a supplemental
carrier in unlicensed spectrum starts to attract some attention in the industry
and is under discussion.
3.2.2 Deployment Scenarios of WiFi
The main WiFi deployment scenario is residential deployment, e.g. to provide
the extension of wireline wideband for the local fixed access from home,
which mainly contains one or few APs and STAs, low cost requirement, and
with low QoS requirement.
In recent years, due to limited spectrum resources, some operators started
to deploy Carrier Grade WiFi (CGW) as a complement to their macro cellular
network, for offloading cellular traffic in some hotspot areas. Generally the
telecommunication level WiFi AP is much more expensive than a residential
AP, and also supports interworking with 2G/3G or 4G networks.
To provide a convenient access to internet, e.g. wireless city project, some
third party or government may also deploy WiFi access in hotspot areas, some
of which adopt telecommunication level AP to support continuous coverage
of the hotspot area and to permit access of more users than residential
networks. However, it is still hard to guarantee the security and QoS.
11
3.3 Commercial Maturity Analysis
From the commercial scale point of view, both LTE and WiFi are very
successful. Both 3GPP and WiFi alliance are paying attention to the
interworking between WiFi and cellular network as well, while some latest
specification requires upgrade of terminal implementation and test. The
industry chain of both LTE small cell and WiFi are analyzed as follows.
Chipset and Terminal Maturity:
LTE protocol phase I (LTE Rel-8) and WiFi phase II (IEEE802.11n) already • released a lot of commercial chipsets and terminals. LTE FDD and TDD
specification are commonly defined, which makes it easy for chipset
design and terminal implementation to support both modes.
WiFi phase III (IEEE802.11ac) is in the roadmap of many vendors, similar • as LTE small cell enhancement (e.g. Rel-11/12).
Candidate spectrum:
LTE small cell may be deployed in the same spectrum as macro cells, and • may also have some small cell specific spectrum. The first batch contains
bands that are already assigned to LTE system, e.g. 2.3GHz and 2.6GHz.
The second batch consists of potential small cell specific bands, e.g.
3.5GHz or spectrums with a bit higher frequency. There is discussion for
potential application in the 5.8GHz band as well.
WiFi residential spectrum is mainly in 2.4GHz, which is already very • crowded. The second batch is around the 5GHz area. The third batch is
60GHz and some potential bands below 1GHz, e.g. TV white space.
12
4. Conclusion
This paper provides the analysis of the small cell solutions based on both LTE
and WiFi, with a focus on the user experience test.
LTE Small Cell is suitable for the typical service application in the cellular
network deployed by an operator or a large scale enterprise. With close
coupling with the Macro layer, the operator can easily manage LTE Small Cell
and control the user experience with the unified interface of the traditional
LTE system, and support good mobility performance. The LTE air interface is
well designed for supporting a substantial number users, guaranteed security
and QoS in diverse deployment scenarios, and complex traffic situations.
WiFi is most suitable for application in residential or small coverage
deployment with a smaller number of users, such as the home, isolated hot
spot or small scale enterprise. The contention based scheduling and network
management of high user density scenario is the bottleneck for WiFi. To
improve its performance in the cellular scenario, higher cost is expected.
13
5. References
GSMA Mobile World congress 2010 Daily, Huawei promises cheaper,
faster broadband, February 2010.
Softbank, 3GPP Rel-12 workshop, June 2012.
Mobile broadband access at home: Informa Telecoms&Media.
TR36.932, Scenarios and requirements for small cell enhancements for
E-UTRA and E-UTRAN.
Huawei, 3GPP Rel12 workshop, June 2012.
IEEE Standard for Information Technology Telecommunications and
information exchange between systems Local and metropolitan area
networks Specific requirements Part 11: Wireless LAN Medium Access
Control (MAC) and Physical Layer (PHY) Specifications, November 2011.
RP-132073, WID Small cell enhancement – Physical layer aspect, 3GPP
RAN#62, Dec. 2013.
TR36.872, Small cell enhancements for E-UTRA and E-UTRAN - Physical
layer aspects.
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
Copyright © Huawei Technologies Co., Ltd. 2013. All rights reserved.
No part of this document may be reproduced or transmitted in any form or by any means without prior written consent of Huawei Technologies Co., Ltd.
Trademark Notice
, HUAWEI, and are trademarks or registered trademarks of Huawei Technologies Co., Ltd.
Other trademarks, product, service and company names mentioned are the property of their respective owners.
HUAWEI TECHNOLOGIES CO., LTD.
Huawei Industrial Base
Bantian Longgang
Shenzhen 518129, P.R. China
Tel: +86-755-28780808
Version No.: M3-024104-20131113-C-1.0
www.huawei.com
General Disclaimer
THE INFORMATION IN THIS DOCUMENT MAY CONTAIN PREDICTIVE STATEMENTS
INCLUDING, WITHOUT LIMITATION, STATEMENTS REGARDING THE FUTURE FINANCIAL
AND OPERATING RESULTS, FUTURE PRODUCT PORTFOLIO, NEW TECHNOLOGY, ETC.
THERE ARE A NUMBER OF FACTORS THAT COULD CAUSE ACTUAL RESULTS AND
DEVELOPMENTS TO DIFFER MATERIALLY FROM THOSE EXPRESSED OR IMPLIED IN THE
PREDICTIVE STATEMENTS. THEREFORE, SUCH INFORMATION IS PROVIDED FOR REFERENCE
PURPOSE ONLY AND CONSTITUTES NEITHER AN OFFER NOR AN ACCEPTANCE. HUAWEI
MAY CHANGE THE INFORMATION AT ANY TIME WITHOUT NOTICE.